VCTUBE is open-source Python library, that can automatically generate
VCTUBE is open-source Python library, that can automatically generate
Recent studies have shown that Text-to-Speech (TTS) systems based on deep neural networks (e.g.,
Tacotron, Deep Voice, etc.) can generate human-like speech with high quality.
However, it has been reported that training such a deep learning model to generate human-like speech
requires a large amount of speech data.
At least 10 hours of
1 | pip3 install vctube | cs |
1 2 3 4 5 6 7 8 9 10 | from vctube import VCtube playlist_name = "" playlist_url = "" lang = "" # ex) ko, en, fr, de ... vc = VCtube(playlist_name, playlist_url, lang) vc.download_audio() #download audios from youtube vc.download_captions() #download captions from youtube vc.audio_split() #split audio with captions | cs |
1 2 3 4 5 | from vctube import VCtube playlist_url = "https://www.youtube.com/watch?v=fj5BcN6Blks" playlist_name="TEST" lang = "en" #ex) ko, en, fr, de... vc = VCtube(playlist_name, playlist_url, lang) | cs |
1 2 3 | vc.download_audio() vc.download_captions() vc.audio_split() | cs |